DESCRIPTION: (Applicant's Abstract) We propose to develop, implement and evaluate a Psychomedical Outreach Intervention (POI) integrated within a primary health care system under the new policy of managed care and aimed at helping injection drug users (IDUs) increase utilization of health care, including drug treatment, reduce HIV risk behaviors and drug use, and improve their health status. The proposed study is a prospective randomized trial comparing an experimental intervention (POI) with a Standard Outreach Intervention. The proposed POI is an integrative community and health care/drug treatment services model. The focus will be on an enhanced outreach strategy and psychomedical support services where outreach workers serve as role models, mediators/brokers between participants and psychomedical services, and motivation agents to help participants accept the patient/client role. Psychiatric nurses trained in motivational interviewing will focus their effort on changing risk behaviors, including drug use, and entering health care services, including drug treatment. Motivational Interviewing will be the approach to the change process, and will be guided by Prochaska and DiClemente's Stages of Change Process Model. A total of 600 IDUs will be recruited from communities of the Arecibo Health Region and randomly assigned to either intervention arm. The recruitment communities will be selected based on median family income and the existence of copping areas. The sampling design follows a 3-stage random selection process of recruitment sites, day and time of recruitment, and selection of a potential participant on the designated site, day and time. For the principal analysis we will focus on comparing the various outcome measures across the two intervention groups. If the preliminary analyses detect a covariate which is not balanced across the two intervention groups, its effect may be statistically adjusted by incorporating the covariate into a logistic or multiple regression model. The second step in the principal analysis will be to consider the "mediating effects" in our design. In this analysis, we wish to discover whether intervention of the experimental arm corresponds to the actual effect of the intervention. Finally, cost-effectiveness analyses will be conducted. For each of the stated outcomes a cost will be estimated. The total cost of resources use will be estimated using claims data from managed care organizations providing services in the area.